Summary The mitotic spindle segregates chromosomes prior to eukaryotic cell division. Motor proteins, crosslinkers, and associated proteins organize spindle microtubules to assemble the bipolar spindle. Among spindle proteins, kinesin-5 motors are particularly important because they are essential to establish a bipolar spindle in most or- ganisms. Tetrameric kinesin-5s are known to crosslink overlapping antiparallel microtubules in the center of the spindle, then step toward plus ends of each microtubule in the pair. This generates force to slide apart antipar- allel microtubules and thereby separate mitotic spindle poles. Contradicting this model are observations that kinesin-5s traf?c in both directions along microtubules and localize near microtubule minus ends, in part by binding -tubulin. Both of these poorly understood mechanisms may be important for spindle-pole separation to establish a bipolar spindle, but whether they are essential is unknown. Additionally, kinesin-5 C-terminal tails are a phosphorylation hotspot important for spindle assembly, but the properties altered by phosphoryla- tion are unclear. This evidence points to a key gap in our understanding of spindle-assembly mechanisms. In preliminary work, we made two key observations: ?rst, kinesin-5/Cut7 moves bidirectionally on ?ssion- yeast spindle microtubules at speeds similar to those measured in vitro. Second, mitotic-spindle-pole separation can occur when Cut7 remains localized at one spindle pole. Thus kinesin-5's two unusual properties may play key roles in positioning the motor to enable proper force generation for spindle pole separation. Further, phos- phorylation of the C-terminal tails may regulate these functions. We hypothesize that kinesin-5 bidirectional movement and spindle-pole localization enable force generation to separate spindle poles, and that cells regu- late this motor's directionality and localization. To test these hypotheses, we will use an interdisciplinary approach combining kinesin-5 perturbation dur- ing mitosis in ?ssion yeast, quantitative light microscopy and image analysis, and computational modeling. To determine the cellular cues that bias kinesin-5/Cut7 directionality toward either plus or minus ends of spindle microtubules, we will test the hypotheses that (1) crowding on the microtubule lattice by binding proteins, (2) motor crosslinking state, and/or (3) phosphorylation alter Cut7 directional bias in vivo. The result will be a sys- tematic comparison of proposed handles cells might use to direct Cut7 on the spindle. To identify the relative contributions of kinesin-5/Cut7 spindle-pole tethering and bidirectional motility to spindle-pole separation, we will measure and model spindle assembly and kinesin-5 localization in cells with speci?c perturbations to motor spindle-pole tethering and directional bias. Further, we will de?ne the role of the Cut7 tail and its phos- phorylation in spindle-pole tethering. This work will systematically test the hypotheses that direct spindle-pole binding and/or bidirectional motility are essential for spindle assembly. The results of this project will de?ne the kinesin-5 properties required for spindle-pole separation during mitotic spindle assembly.